In a network computing environment, many actions take place automatically for multiple computers during scheduled time periods. For example, these actions include installing software applications, upgrades, and patches, running virus scan programs, routine maintenance, and/or other scheduled actions. Typically, these actions are scheduled to start at a specific time.
Various problems may occur when a large number of computers access resources to perform an action at the same time. For example, a high network load may occur resulting in network congestion. Other problems may also occur.
Some systems have attempted to overcome these problems by specifying a time period for the action and enabling computers or users to select a “random” time within the specified time period for the action to be performed. However, in such cases, the “random” mechanism must be invoked on each individual computer and therefore computers can still end up performing the action at or about the same time. Furthermore, due to the randomized start times, the behavior of the computers on the network may become unpredictable depending on the time period.
These and other drawbacks exist.